Connect with us

Ethereum

Ethereum [ETH]: ProgPow audit may not be completed before Istanbul

Priya

Published

on

Ethereum [ETH]: ProgPow audit may not be completed before Istanbul
Source: Unsplash

Programmatic Proof-of-Work [ProgPow] has undoubtedly been one of the most controversial upgrades of Ethereum. The algorithm that promises to ward off ASIC mining would probably not make it till after the Istanbul hardfork. A blog post titled ‘ProgPow Audit Delay Issue’ published by Hudson Jameson, a member of EF, stated that they encountered problems with ProgPow audit.

Hudson Jameson explained the situation,

“We had a hardware partner who specialized in ASICs who was going to work with Least Authority to perform the hardware parts of the audit. They are no longer participating in the audit so we are looking for other auditors for the hardware portion.”

Further, Jameson stated that there were “some good candidates” to fill in the position, however, it would “effectively delay” the start of the audit even further than what they had anticipated. This was the sole reason the team was uncertain whether the audit would be ready before the Istanbul hardfork.

Along with this, the member also spoke about the funding pertaining to ProgPow miner, where he remarked,

“On top of that I am not sure if anyone has sorted the funding situation in order to build an open source ProgPoW miner.”

Due to these problems, Hudson presented two options, first, delay “ProgPoW until the hardfork after Istanbul”, and second, let ProgPow have its own hardfork once the audit was completed. Hudson further stated that “this was not an ideal situation at all, but despite our best efforts, it is what we have before us”.

McDogger, a Redditor said,

“I personally would prefer a third option (stop with the audits, drop progpow and spend time/ money on Eth 1.x finality) but if I had to chose than it’s certainly option 1. Progpow […] doesn’t deserve it’s own – contentious, because it’s just one change and not many other improvements – hardfork.”

The initial announcement was made on Ethereum Cat Herders Update #9, where Pooja Ranjan, a EF member, stated that “logistical issues” were the reason for the delay of ProgPow audit, adding that the timeline for the start and end of the audit were not even decided yet.





Subscribe to AMBCrypto’s Newsletter




Follow us on Telegram | Twitter | Facebook



Priya is a full-time member of the reporting team at AMBCrypto. She is a finance major with one year of writing experience. She has not held any value in Bitcoin or other currencies.

News

Coin Metrics data reveals inaccuracies in Kik’s claim of being as dominant as BTC, ETH blockchains

Avatar

Published

on

Source: Pixabay

Upon investigating Kik’s claims in response to SEC’s lawsuit filed earlier this month, CoinMetric data reported inconsistencies in the on-chain activity and adoption rate of its native token, Kin.

In a study dubbed, “An Analysis of Kin’s On-Chain Activity,” the crypto-asset elaborated on the two assertions made by Kik in its letter to the US Securities and Exchange Commission.

Kik’s first claim was regarding its blockchain activity. Its in-house token, Kin, supposedly exceeded Ether and Bitcoin to record the fifth highest daily blockchain activity. This was debunked by CoinMetric’s investigation after taking into account its “Operation Count” [the same metric used by Kik to support their claim] and “Transfer Value.”

In terms of the Operation Count, the report explained,

“According to Kik’s source for the metric, “blockchain activity” is defined as “the number of operations on the blockchain in the last 24 hours.” Operations are broadly defined as any type of action that could be recorded on chain. But operations are not standardized across blockchains which makes comparing across chains difficult.”

Besides, drawing parallel comparisons across blockchains with radically different use cases and operations is difficult.

Although Kik’s original research showed a high number of account creations, Coin Metrics data revealed that many of these accounts were empty.

Additionally, Kin’s “create account” operation has a fee of .001 Kin. The report highlighted that a metric such as “operations count” for the purpose of blockchain activity cannot be used as a measurement tool since Bitcoin and Ethereum blockchains do not track account creations on-chain.

In terms of Transfer of Value, the report elaborated,

“Theoretically, high daily transfer value should signify high activity. But transfer value is often quite noisy, especially on low fee blockchains where there are minimal costs to sending transactions. Some transfers might simply be users moving money around between addresses they own”

Instead, Coin Metrics contrived “adjusted transfer value” metric to eliminate what it called, “noise and certain artifacts like self-sends, or deliberate spammy behavior.” Coin Metric noted that this gives a clearer picture of the on-chain activity, resulting in a decreased transfer value when compared to other blockchains, even if it had a high number of daily blockchain operations.

Additionally, Kin’s average transaction value was also low, when compared to other blockchains. For the first claim, Coin Metrics concluded that the Kin platform had more micro-transactions than Bitcoin and other dominant blockchains, while highlighting the fact that the latter blockchains are not primarily used for such transactions.

Regarding Kik’s second claim that said that over 300,000 users were earning and spending Kin as a currency, Coin Metrics assessed its blockchain usage. The number of addresses is not necessarily equal to the number of users since a single user could have multiple addresses. Hence, Coin Metrics took the number of active users into account, which the report defined as “the number of unique addresses that were active in the network [either as a recipient or originator of a ledger change] during that day.” The report noted,

“Kin 2 has significantly more originating active addresses than Kin 3. Although Kin is in the process of migrating to Kin 3, it appears that Kik is using data from the Kin 2 chain to support their claims about usage.”

Further, Kin 2 and Kin 3 had more active addresses that received payments than originated payments, which meant that there were more “earners” on Kin than “spenders,” also noting that only 35,000 addresses held over 10,000 kin [nearly $0.23]. The report added that the figures are lower than other blockchains which have a minimum of 1,000,000 addresses with at least $1.

After examining multiple critical aspects, Coin Metrics concluded that Kin fell below dominant blockchains in terms of daily active addresses, despite maintaining steady growth. It said,

“A majority of Kin’s active addresses have small account balances. While this makes sense for a network built around micropayments, when viewed across multiple metrics, our data show that Kin is not more widely used than dominant chains such as Bitcoin or Ethereum.”





Subscribe to AMBCrypto’s Newsletter


Continue Reading